Summary

Overview

Isoform classification

Performance metrics

Simulated transcripts were categorized into known simulated transcripts (FSM), which are present in the reference annotation GTF, and novel simulated transcripts that were excluded from the annotation by SQANTI-SIM. The following metrics were computed:

  • Number of simulated isoforms: Total number of simulated isoforms.
  • True Positive isoforms (TP): Isoforms that match junctions with a simulated transcript, and both their TSS and TTS fall within the 50bp threshold.
  • Partial True Positive isoforms (PTP): Isoforms that match junctions, but either the TSS or TTS is beyond the threshold.
  • False Negative isoforms (FN): Isoforms that were simulated but not detected.
  • False Positive isoforms (FP): Transcripts that were detected but were not part of the simulated set.
  • Sensitivity: TP / Number of simulated transcripts.
  • Precision: TP / Detected isoforms.
  • F-score: The harmonic mean of precision and sensitivity.
  • False Discovery Rate: (FP + PTP) / Detected isoforms.
  • Positive Detection Rate: (TP + PTP) / (TP + FN).
  • False Detection Rate: FP / (TP + PTP + FP).

Tables

Full results

Above min. supp. results

Plots

Compact

Radar chart

Isoform characterization

Isoform distribution

Simulated reads

Number of exons

Transcript length

Isoform complexity

Splice junctions

Orthogonal data support

Short reads

CAGE peaks

Short reads + CAGE peaks